NO. lO STORM FREQUENCY IN THE UNITED STATES KULLMER 3 



of a relation to sunspot latitudes. Gone entirely was the pattern 

 with the southerly projection, which I had found consistently for 

 six solar cycles. This pattern still remains an enigma, the result of 

 some solar relationship, the difference between years of maximum 

 solar activity and those of minimum activity. Instead I discovered 

 a solar relationship of possibly great significance. 



The first period, figure 6, shows three horizontal bands. Plus 

 figures indicate more storms at the beginning of the cycle ; minus 

 figures, more storms at the end of the cycle than at the beginning. 

 The following period, figure 7, shows a map of an entirely different 

 character — a plus band in the north, and curving down the Atlantic 

 coast and across the south a minus area, forming a script T pattern. 

 The succeeding period, figure 8, returns to the pattern of the first, 

 three horizontal bands, but in reversed order, with a pronounced 

 minus area between the plus bands. The succeeding map, figure 9, 

 was the greatest surprise of all — a return to the almost identical 

 script T pattern of the second period, but in reverse, with a minus 

 area in the north and curving down the Atlantic coast, and across the 

 south a plus area. And now with the -fifth period, figure 10, we re- 

 turn to the three horizontal bands, but again in reverse, with a plus 

 area between two still widened minus areas. These complicated re- 

 lationships will appear more simple in a table, figure 11. Because 

 of the appearance of three bands in the first and third period and 

 the strange script T pattern in the second and fourth periods, we 

 may divide the first four periods into two Hale solar cycles. But 

 whatever the solar influence may be that determines the location 

 of storms on earth, that influence was completely reversed in the two 

 Hale cycles. The evidence of a reversal is further supported by the 

 return in the fifth period to the conditions found in the first. 



A crucial test will be oft'ered by the coming sixth solar cycle, which 

 will have to show the script T pattern with a plus area in the north 

 and the long curved minus area in the south. Fortunately we arc 

 able to look somewhat into the future. I have combined the maps for 

 1935- 1937, and they show the highest known figures for the eastern 

 Canadian region. It might be considered that this shows merely the 

 increase in the network of northern Canadian recording stations. 

 But I present the evidence of the two squares in Quebec, north of 

 the Gaspe. Figure 12 shows the number of barometric depressions 

 that crossed these two squares from 1883 to 1940. This diagram is 

 another evidence that tends to confirm my confidence in the early 

 records of tracks of storms. The early years show approximately 

 the same high frequency as the later years; the year 1894 with 56 



